© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126670
    Prev
    Next
    INDEX
    Explanations

    model performance

    np_max-act · gemini-2.0-flash

    This neuron activates on words and phrases that talk about improving or measuring performance (e.g., “improve,” “performance,” “improving”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    áno
    -0.06
     Thumbnail
    -0.06
     zkou
    -0.06
    anlık
    -0.06
     salad
    -0.06
     estudio
    -0.05
     simulator
    -0.05
     giảng
    -0.05
    unar
    -0.05
    CDF
    -0.05
    POSITIVE LOGITS
     ''),
    0.09
    ceil
    0.07
    ={},
    0.07
     声
    0.07
     microseconds
    0.07
     alg
    0.07
     competing
    0.07
    Proceed
    0.07
    (hero
    0.06
     Raises
    0.06
    Activations Density 0.033%

    No Known Activations